Skip to main content

Active learning for online bayesian matrix factorization

Publication ,  Journal Article
Silva, J; Carin, L
Published in: Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
September 14, 2012

The problem of large-scale online matrix completion is addressed via a Bayesian approach. The proposed method learns a factor analysis (FA) model for large matrices, based on a small number of observed matrix elements, and leverages the statistical model to actively select which new matrix entries/observations would be most informative if they could be acquired, to improve the model; the model inference and active learning are performed in an online setting. In the context of online learning, a greedy, fast and provably near-optimal algorithm is employed to sequentially maximize the mutual information between past and future observations, taking advantage of submodularity properties. Additionally, a simpler procedure, which directly uses the posterior parameters learned by the Bayesian approach, is shown to achieve slightly lower estimation quality, with far less computational effort. Inference is performed using a computationally efficient online variational Bayes (VB) procedure. Competitive results are obtained in a very large collaborative filtering problem, namely the Yahoo! Music ratings dataset. © 2012 ACM.

Duke Scholars

Published In

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

DOI

Publication Date

September 14, 2012

Start / End Page

325 / 333
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Silva, J., & Carin, L. (2012). Active learning for online bayesian matrix factorization. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 325–333. https://doi.org/10.1145/2339530.2339584
Silva, J., and L. Carin. “Active learning for online bayesian matrix factorization.” Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, September 14, 2012, 325–33. https://doi.org/10.1145/2339530.2339584.
Silva J, Carin L. Active learning for online bayesian matrix factorization. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012 Sep 14;325–33.
Silva, J., and L. Carin. “Active learning for online bayesian matrix factorization.” Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sept. 2012, pp. 325–33. Scopus, doi:10.1145/2339530.2339584.
Silva J, Carin L. Active learning for online bayesian matrix factorization. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2012 Sep 14;325–333.

Published In

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

DOI

Publication Date

September 14, 2012

Start / End Page

325 / 333